Odds Modelling and Testing Inefficiency of Sports Bookmakers

Summary during Conducting the Project

Apr-2008 to Apr-2010 Odds Modelling and Testing Inefficiency of Sports-Bookmakers

  • Learn RExcel, CrystalBall, ModelRisk etc. and choose R open source software to start my research.
  • Collect the livescore and also 1x2, Asian Handicap, Over Under odds price data of 29 sportsbookmakers manually from 500WAN, BET007 and NowGoal website and filter the odds price data from 2006 to 2011.
  • Apply Poisson model in R to test the return of the investment. This research job is the most completed, success and the first research which write the whole odds compilation EM model and data management by refer to thousands of research papers in sportsbook odds modelling after resigned from Caspo Inc.

Similar with fbRanks r package where I used to simulate few years ago via Dixon & Coles 1996 by englianhu (2014). You might refer to the author’s blog LastPlanet Soccer Ranking. Meanwhile, my package includes not only the 1x2 but all possible products includes :

  • 1x2 (Both 1st-Half and Full Time)
  • Asian Handicap (Both 1st-Half and Full Time)
  • Over Under (Both 1st-Half and Full Time)
  • Correct Score (Both 1st-Half and Full Time)
  • Odd Even (Both 1st-Half and Full Time)
  • Half-Time/Full-Time (Both 1st-Half and Full Time)

There are some functions from modelling, database management to staking. You are feel free to browse over source code via Rmodel r package. However, I wish to refer to QuantTools r package and write a real-time price trends database for quantitative trading.

if(!require('devtools') install.packages('devtools'))
devtools::install_github('englianhu/Rmodel')

1. Abstract

In this paper I am applied a diagonal inflated biviriate poisson as well as a simple staking model whereby evaluate the efficiency of odds price of Asian Handicap and Goal Line offered by 40 sports bookmakers. Finally I get a breakdown profit & lose table. While I used Kelly model1 next to this research which generated profit (positive return of investment) more than 30% every year.

2. Full Version Thesis

2.1 Odds Modelling

The research on the soccer odds modelling, result prediction, staking as well as the return of investment is applicable to real life. Kindly refer to Odds Modelling and Testing Inefficiency of Sports-Bookmakers.pdf to view the paper.

Kindly refer to Odds Modelling and Testing Inefficiency of Sports-Bookmakers to read the embed online read mode pdf version.

2.2 Betting Strategy

Due to my previous research applied $1 as long as the edge of EM2 odds is over BK3 odds and concludes that the staking methods need to be improved.

Here I tried to scrap the odds price from 7M and NowGoal.com website4, and apply Kelly-Criterion Model, from the simulatioin we can know that the EM model is profitable.

3. Odds Modelling Version II

In my previous Betting-Strategy-and-Model-Validation, I enhanced my Rmodel and test the return of investment. Here I collect the odds price trends of bookmakers and directly fit into calculation as refer to Niko (2006).

Gianluca Baio & Marta Blangiardo (2010) introduced a model which is not inferior to the one used by Karlis & Ntzoufras (2003). Ioannis Ntzoufras (2009) also using WinBugs for modelling where it (and OpenBugs) are not user friedly.

3.1 Mixed Model

…

3.2 Betting Strategy

…

  1. Refer to Testing Inefficiency of Sports-Bookmakers by Kelly Model↩︎

  2. Expectation Maximization↩︎

  3. Odds price offer by Bookmakers↩︎

  4. You are feel free to read from WebDriver-DynamicWebpage-Scrapping.↩︎